REVIEW 2 major objections 4 minor 15 references
Paradoxical decompositions of finite-dimensional non-Archimedean normed spaces
T0 review · 2 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read Every finite-dimensional non-Archimedean normed space of dimension at least two is paradoxical using four affine isometry pieces.
desk verdict Solid advancement on four-piece paradoxes in non-Archimedean normed spaces; the mixed-characteristic proof needs a stated and verified external theorem. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The mechanism is the special affine congruence subgroup $SA(n,D_K,\varepsilon)$: affine maps whose linear part lies close to the identity matrix, has determinant one, and has entries in the valuation ring $D_K$, with a small translation part. Proposition 3.1 shows that whenever a norm is equivalent to the maximum norm, this group lies inside the affine isometry group. Theorem 3.9 then shows that for every $n\ge2$ and every $\varepsilon>0$, $SA(n,D_K,\varepsilon)$ contains a free subgroup of rank two acting locally commutatively on $K^n$, meaning every point's stabilizer is Abelian. The construction splits by characteristic: in the unequal-characteristic case it combines an explicit free subgroup of $SL(2,\mathbb{Z})$ with a uniform freeness theorem to obtain short free elements; in the equal-characteristic case it builds explicit generators from a diagonal matrix $a=\operatorname{diag}(\lambda,\lambda^{-1})$ and a conjugate $b=hah^{-1}$ satisfying the ping-pong lemma. Lemma 3.3 embeds the two-dimensional construction into higher dimensions, and the bridge from a free locally commutative subgroup to a four-piece paradox is the standard characterization in [15].
What would settle it
Verify the imported uniform freeness theorem on the specific group generated by $g=(A_{p^s},(2p^s,0))$ and $h=(A_{p^t},(2p^s,0))$ in $SA(2,\mathbb{Z},\varepsilon)$: if for some prime $p$ and integers $s<t$ the theorem's hypotheses fail, or no product of at most $N$ generators contains two elements generating a free subgroup acting locally commutatively on $\mathbb{Q}^2$, then the unequal-characteristic branch of Proposition 3.4, and with it Theorem 3.10 in the unequal-characteristic case, has no supporting proof.
Extended reading notes
Core claim
The central result is Theorem 3.10: if $(K,|\cdot|)$ is a field with a nontrivial non-Archimedean valuation, $n\ge 2$, and $\|\cdot\|$ is any non-Archimedean norm on $K^n$ equivalent to the maximum norm, then $K^n$, every closed ball $B[x_0,r]$, every open ball $B(x_0,r)$, and every nonempty sphere $S[x_0,r]$ are paradoxical with respect to the group of affine isometries of $(K^n,\|\cdot\|)$ using four pieces. Because over a complete nontrivially valued field every finite-dimensional normed space is linearly isometric to one of these and all norms are equivalent, the conclusion extends to every finite-dimensional non-Archimedean normed space of dimension at least two.
Load-bearing premise
The proof's unequal-characteristic branch depends on an imported theorem guaranteeing that a certain free group contains short elements generating a free subgroup with locally commutative action; the theorem's exact hypotheses are not checked in the paper.
Editorial extensions
If this is right
- Every finite-dimensional non-Archimedean normed space of dimension at least two over a complete nontrivially valued field is four-piece paradoxical, including all of its balls and spheres (Corollary 3.11).
- Four is the optimal number of pieces: the four-piece criterion in [15] rules out any paradoxical decomposition with fewer pieces.
- The result covers discrete and dense valuations uniformly, and it holds in both zero and positive characteristic whenever the valuation is nontrivial.
- For trivially valued non-locally-finite fields, paradoxicality is decided by the norm's value set: fewer than $n$ nonzero values gives a four-piece paradox, while $n$ distinct values gives none (Theorem 3.13).
- Locally finite fields and the one-dimensional line over any field remain non-paradoxical even for the full affine group (Theorem 3.14).
Reading between the lines
- Beyond the paper's existence statements, the explicit ping-pong sets in Lemma 3.6 are described by concrete valuation inequalities, so the four-piece decompositions in the equal-characteristic branch could in principle be made algorithmic rather than purely existential.
- The result points to a sharp contrast with Archimedean geometry: no non-Archimedean norm equivalent to the maximum norm can hide its isometries the way Archimedean norms can, so paradoxicality is the default behavior rather than a special construction.
- One natural extension the author does not pursue is to replace affine isometries by locally affine or piecewise isometric maps; the same free locally commutative subgroup would likely transfer the paradox, but the paper does not address such weakened groups.
- The trivially valued dichotomy in Theorem 3.13 suggests a testable principle: the number of distinct nonzero norm values controls whether a finite-dimensional space is paradoxical, and it is natural to ask whether a similar value-set criterion governs paradoxicality for non-trivially valued fields as well.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies paradoxical decompositions in finite-dimensional non-Archimedean normed spaces. Its main theorem (Theorem 3.10) states that if (K,|·|) is a nontrivially valued non-Archimedean field, n ≥ 2, and ||·|| is a non-Archimedean norm on K^n equivalent to the maximum norm, then K^n, every closed ball, every open ball, and every nonempty sphere are paradoxical with respect to the affine isometry group of (K^n,||·||) using four pieces. The proof first shows (Proposition 3.1) that a suitable principal congruence subgroup of the affine group acts by isometries, then proves (Theorem 3.9) that this congruence subgroup contains a free rank-two subgroup acting locally commutatively, using two separate characteristic cases (Proposition 3.4 for mixed characteristic, Proposition 3.7 for equal characteristic), and finally applies the Tomkowicz–Wagon criterion. Corollary 3.11 extends the result to all finite-dimensional normed spaces over complete fields. The paper also contains a necessary-and-sufficient result in the trivially valued non-locally-finite case (Theorem 3.13) and negative results for locally finite fields and dimension one (Theorem 3.14).
Significance. If the issues below are resolved, this is a valuable contribution. It improves the earlier five-piece results in [8] to the optimal four pieces and extends them from the maximum norm to all equivalent norms, which is a genuinely non-Archimedean phenomenon: in the Archimedean setting every Banach space has an equivalent norm with only trivial isometries (Jarosz). The equal-characteristic construction is explicit and self-contained (Lemma 3.6 and Proposition 3.7), the reduction from arbitrary norms to congruence subgroups is clean, and the negative results are simple but useful. The mixed-characteristic branch depends on an external theorem of Pham, but this is a dependency, not circularity: there are no fitted parameters, and the central claim is not derived from the target result.
major comments (2)
- [§3.2, Proposition 3.4] The proof applies [10, Theorem 1.1] to the group Gamma = ⟨g,h⟩ ≤ SA(2,Z,epsilon) without stating the theorem's hypotheses. This application is the only non-self-contained step in the mixed-characteristic branch, and that branch is needed for fields such as Q_p. Please state the hypotheses of [10, Theorem 1.1] and verify that Gamma and the generating set S satisfy them, in particular any conditions on the ambient group, the finite generating set, or the action on Q^2. If the theorem has additional hypotheses not verified in the text, the proof of Theorem 3.9 is incomplete for all mixed-characteristic fields.
- [§3.2, Lemma 3.6] The verification of condition (3.3) of Lemma 3.5 is incomplete as written. The displayed inclusion h(U_a^- ∩ U_{a^{-1}}^-) ∪ h^{-1}(U_a^- ∩ U_{a^{-1}}^-) ⊆ (K^2 minus U_a^-) ∩ (K^2 minus U_{a^{-1}}^-) only rules out triples of sets containing both U_a^- and U_{a^{-1}}^-. The required disjointness for the triples {a,b,b^{-1}} and {a^{-1},b,b^{-1}} is not established by that inclusion. Since condition (3.3) is one of the hypotheses of the ping-pong lemma used to obtain the free subgroup, this leaves the self-contained equal-characteristic construction incomplete. Please supply the missing case analysis or show explicitly how the stated inclusion implies the remaining cases.
minor comments (4)
- [§3.1, Lemma 3.3] In the statement and proof, the containment should read ι(SA(2,D_K,epsilon)) ≤ SA(n,D_K,epsilon), not ι(SA(2,D_K,epsilon)) ≤ ι(SA(n,D_K,epsilon)).
- [§2.1] There is a typo in 'We us also define'; it should be 'We also define'.
- [§3.2, Proposition 3.4] The sentence 'It is easy to see that any x in K^2 outside Q^2 has the trivial stabilizer in Gamma' would benefit from a one-line justification: since every non-identity element of Gamma has a unique fixed point in Q^2, no point outside Q^2 can be fixed by a non-identity element.
- [§3.2, Proposition 3.7] The proof uses an auxiliary valuation |·|_1 to run the ping-pong argument; this is legitimate because local commutativity is an algebraic property of the action, but the text should explicitly say that the group elements constructed remain in SA(2,D_K,epsilon) for the original valuation and that the auxiliary valuation is used only to establish the free group property.
Circularity Check
No circularity: the proof uses external theorems (Magnus, Pham, Tomkowicz–Wagon, Schikhof) and the only self-citation [8] is motivational.
full rationale
I traced the derivation chain. The main theorem 3.10 depends on Theorem 3.9, which combines Proposition 3.4 (mixed characteristic) and Proposition 3.7 (equal characteristic). Proposition 3.7 is fully self-contained: it constructs explicit generators a,b and verifies the ping-pong Lemma 3.5 using Lemma 3.6; no input is reused as output. Proposition 3.4 invokes Magnus's basis of a free subgroup of SL(2,Z) and Pham's Theorem 1.1 to pass from a free non-virtually-solvable group Gamma to a free rank-2 subgroup acting locally commutatively on Q^2; this is an external result, not a restatement of the paper's target, and the paper does not fit any parameter to the conclusion. The equivalence between four-piece paradoxical decompositions and free locally-commutative rank-two actions is quoted from Tomkowicz–Wagon, again external. The only reference to the author's earlier paper [8] occurs in the introduction as background/motivation; no proof step invokes [8]. The potential concern that Pham's theorem's hypotheses are not stated is a completeness or correctness risk, not a circularity, since circularity would require the cited theorem to encode the paper's own conclusion by construction. No fitted inputs, no self-definitional quantities, and no renamings of known results occur.
Assumptions & free parameters
assumptions (6)
- standard math Four-piece paradoxical decomposition is equivalent to the existence of a rank-two free subgroup acting locally commutatively ([15, Theorems 5.5 and 5.8]).
- standard math Pham's uniform Tits alternative [10, Theorem 1.1] gives, in a bounded word ball of a non-virtually-solvable affine group over Q, two elements generating a free group with locally commutative action.
- standard math Magnus's construction [7, Theorems 3 and 4] produces a free non-parabolic subgroup of SL(2,Z) generated by A_{p^s} and A_{p^t}.
- standard math Krull's valuation extension theorem [12, Theorems 14.1 and 14.2], together with the fact that elements of valuation less than 1 are transcendental over a trivially valued prime subfield.
- standard math Amenable groups admit no paradoxical actions ([5, Theorem A.13]); locally finite groups and solvable groups are amenable.
- standard math Over a complete non-Archimedean valued field, every finite-dimensional norm is equivalent to the maximum norm ([12, Theorem 13.3]).
Cite this review
Pith. "Pith review of Paradoxical decompositions of finite-dimensional non-Archimedean normed spaces." pith.science (2026). https://pith.science/paper/HOCK5KGX
@misc{pith2026250601528,
author = {Pith},
title = {Pith review of: Paradoxical decompositions of finite-dimensional non-Archimedean normed spaces},
year = {2026},
howpublished = {\url{https://pith.science/paper/HOCK5KGX}},
note = {Machine review of arXiv:2506.01528}
}
abstract
We show that any normed space $(K^n,\|\cdot\|)$, $n\ge 2$, over a field $K$ equipped with a nontrivial non-Archimedean valuation admits a paradoxical decomposition using four pieces with respect to the group of its affine isometries, provided that the norm $\|\cdot\|$ is equivalent to the maximum norm. It follows that any finite-dimensional normed space $(X,\|\cdot\|)$ with $\dim{X}\ge 2$ over a complete non-Archimedean nontrivially valued field $(K,|\cdot|)$ is paradoxical using four pieces with respect to the group of its affine isometries.
Reference graph
Works this paper leans on
-
[8]
K. Orzechowski , The B anach- T arski paradox for some subsets of finite-dimensional normed spaces over non-archimedean valued fields , 2024, https://arxiv.org/abs/2402.14772
work page Pith review arXiv 2024
-
[1]
S. Banach and A. Tarski , Sur la d \'e composition des ensembles de points en parties respectivement congruentes , Fund. Math., 6 (1924), pp. 244--277
work page 1924
-
[2]
E. Breuillard, B. Green, R. Guralnick, and T. Tao , Expansion in finite simple groups of L ie type , J. Eur. Math. Soc. (JEMS), 17 (2015), pp. 1367--1434
work page 2015
-
[3]
Hausdorff , Grundz \"u ge der Mengenlehre , Veit and Company, Leipzig, 1914
F. Hausdorff , Grundz \"u ge der Mengenlehre , Veit and Company, Leipzig, 1914
work page 1914
-
[4]
Jarosz , Any B anach space has an equivalent norm with trivial isometries , Israel J
K. Jarosz , Any B anach space has an equivalent norm with trivial isometries , Israel J. Math., 64 (1988), pp. 49--56
work page 1988
-
[5]
K. Juschenko , Amenability of discrete groups by examples , American Mathematical Society, Providence, RI, 2022
work page 2022
-
[6]
A. Lubotzky and D. Segal , Subgroup growth , vol. 212 of Progress in Mathematics, Birkh\"auser Verlag, Basel, 2003
work page 2003
-
[7]
W. Magnus , Rational representations of F uchsian groups and non-parabolic subgroups of the modular group , Nachr. Akad. Wiss. G\" o ttingen Math.-Phys. Kl. II, (1973), pp. 179--189
work page 1973
Show all 15 references
-
[9]
Perez-Garcia and W
C. Perez-Garcia and W. H. Schikhof , Locally convex spaces over non- A rchimedean valued fields , Cambridge University Press, Cambridge, 2010
2010
-
[10]
L. L. Pham , Uniform K azhdan constants and paradoxes of the affine plane , Transform. Groups, 27 (2022), pp. 239--269
2022
-
[11]
Sat\^ o , A locally commutative free group acting on the plane , Fund
K. Sat\^ o , A locally commutative free group acting on the plane , Fund. Math., 180 (2003), pp. 25--34
2003
-
[12]
W. H. Schikhof , Ultrametric calculus. An introduction to p -adic analysis , Cambridge University Press, Cambridge, 1984
1984
-
[13]
Schneider , Nonarchimedean functional analysis , Springer Monographs in Mathematics, Springer-Verlag, Berlin, 2002
P. Schneider , Nonarchimedean functional analysis , Springer Monographs in Mathematics, Springer-Verlag, Berlin, 2002
2002
-
[14]
Tits , Free subgroups in linear groups , J
J. Tits , Free subgroups in linear groups , J. Algebra, 20 (1972), pp. 250--270
1972
-
[15]
Tomkowicz and S
G. Tomkowicz and S. Wagon , The B anach- T arski paradox , Cambridge University Press, New York, second edition ed., 2016
2016
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.